The spatial extent of tauopathy on [18F]MK-6240 tau PET shows stronger association with cognitive performances than the standard uptake value ratio in Alzheimer’s disease

• Published in: European journal of nuclear medicine and molecular imaging 2024-05, Vol.51 (6), p.1662-1674
• Main Authors: Gérard, Thomas, Colmant, Lise, Malotaux, Vincent, Salman, Yasmine, Huyghe, Lara, Quenon, Lisa, Dricot, Laurence, Ivanoiu, Adrian, Lhommel, Renaud, Hanseeuw, Bernard
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Alzheimer Disease - diagnostic imaging; Alzheimer's disease; Biological Transport; Biomarkers; Brain; Brain - diagnostic imaging; Brain - metabolism; Cardiology; Cerebrospinal fluid; Cognition; Cognition & reasoning; Cognitive ability; Correlation; Diagnostic systems; Disease; Female; Humans; Imaging; Isoquinolines; Male; Medicine; Medicine & Public Health; Memory; Middle Aged; Missing data; Neurodegenerative diseases; Nuclear Medicine; Oncology; Original; Original Article; Orthopedics; Pathology; Positron-Emission Tomography; Radiology; Radiopharmaceuticals - pharmacokinetics; Sensitivity analysis; Signs and symptoms; Standard scores; Tau protein; tau Proteins - metabolism; Tauopathies - diagnostic imaging
• ISSN: 1619-7070; 1619-7089
• DOI: 10.1007/s00259-024-06603-2

Purpose [ 18 F]MK-6240, a second-generation tau PET tracer, is increasingly used for the detection and the quantification of in vivo cerebral tauopathy in Alzheimer’s disease (AD). Given that neurological symptoms are better explained by the topography rather than by the nature of brain lesions, our study aimed to evaluate whether cognitive impairment would be more closely associated with the spatial extent than with the intensity of tau-PET signal, as measured by the standard uptake value ratio (SUVr). Methods [ 18 F]MK6240 tau-PET data from 82 participants in the AD spectrum were quantified in three different brain regions ( Braak  ≤  2 , Braak  ≤  4 , and Braak  ≤  6 ) using SUVr and the extent of tauopathy (EOT, percentage of voxels with SUVr ≥ 1.3). PET data were first compared between diagnostic categories, and ROC curves were computed to evaluate sensitivity and specificity. PET data were then correlated to cognitive performances and cerebrospinal fluid (CSF) tau values. Results The EOT in the Braak  ≤  2 region provided the highest diagnostic accuracies, distinguishing between amyloid-negative and positive clinically unimpaired individuals (threshold = 9%, sensitivity = 79%, specificity = 82%) as well as between prodromal AD and preclinical AD (threshold = 38%, sensitivity = 81%, specificity = 93%). The EOT better correlated with cognition than SUVr (∆R 2  + 0.08–0.09) with the best correlation observed for EOT in the Braak  ≤  4 region ( R 2  = 0.64). Cognitive performances were more closely associated with PET metrics than with CSF values. Conclusions Quantifying [ 18 F]MK-6240 tau PET in terms of EOT rather than SUVr significantly increases the correlation with cognitive performances. Quantification in the mesiotemporal lobe is the most useful to diagnose preclinical AD or prodromal AD.